A number of presentations at CROI 2017 showed data on early HIV treatment and diagnostics in young children.
Treatment of acute infection in neonates
Just over 30% of infants achieved undetectable viral load and 10% became PCR negative after very early initiation of treatment in South African study. 
ART in primary infection may reduce the size of the viral reservoir and allow viral control off treatment. Infants can be identified soon after infection.
The LEOPARD study tracks the response to ART in neonates started on ART in early infection at Rahima Moosa Mother and Child Hospital in Johannesburg.
Of 30 neonates who started ART within 48 hours of birth with six months or to one year follow up, over half (57%) were male and mean birth weight was 3015 grams. All infants received nevirapine (NVP), lamivudine (3TC) and zidovudine (AZT) with substitution of lopinavir/ritonavir (LPV/r) for nevirapine at median of 27 days (IQR 18 to 32) later.
Approximately 25% of mothers had received no treatment before delivery and 25% <12 weeks of ART; the remainder were already on ART or had received >12 weeks of treatment. Median infant viral load before ART was 30,000 copies/mL (range 60 to >2 million).
Viral load was measured at frequent intervals lower detection limit of 20 copies/mL and qualitative HIV diagnostic PCR tests were repeated.
Three of the 30 infants died at 43, 61 and 89 days respectively; all were male. Only one was low birth weight but all three had high baseline viral load.
There was wide variability in virological response among the remaining infants – from those who did not achieve undetectable viral load to three who became PCR negative. More than a third achieved and sustained undetectable viral load. The children remain in follow up.
Rapid decline of HIV DNA in infants starting very early ART
Infants receiving early ART experience very rapid early phase HIV and DNA decay according to a related study conducted at Stellenbosch University, South Africa. 
The study was designed to investigate total HIV DNA kinetics in infants who started ART as soon as possible after birth.
Eleven infants diagnosed through a public health sector birth diagnosis programme started ART 0 and 8 days after birth (median 3 days). Infants started ART with AZT/3TC/NVP, with NVP replaced by LPV/r after two weeks of age and AZT replaced by abacavir at three months of age.
Peripheral blood mononuclear cells (PBMCs) and plasma were processed at three monthly visits. Total DNA was measured with a sensitive quantitative PCR assay and RNA was also quantified.
Median baseline viral load was 4.0 (range 2.4 to 4.7) log10 copies/mL. Five infants were included in the kinetic study.
In three infants RNA declined to <100 copies/mL within 3.3 months. In the other two this occurred within 6.3 and 6.7 months. DNA decay was in two phases, very fast in the first two weeks, then relatively slow but progressive.
DNA decay in three infants with <100 copies/mL before 3.3 months, phase 1: conditional R2 0.97 (95%CI: 0.90 to 1.00); median decay -2.3 log10 copies/month (range -2.1 to -2.7); 200 fold/month (range 122 to 545). Half-life 4 days (range 3.4 to 4.4).
Phase 2 in five infants: conditional R2 0.97 (95% CI: 0.90 to 1.00); median decay -0.15 log10 copies/month (range -0.13 to -0.2); 65 fold/year (range 33 to 222). Half-life 60.7 days (range 46.9 to 72.4).
The first phase was much faster than that reported in adults. The investigators noted that it was a conservative estimate as baseline DNA could be even higher as they divided by total cells in dried blood spots and only about 50% were lymphocytes. Second phase decay is much faster than in infants starting ART at two months.
Very rapid early phase HIV RNA and DNA decay poses a diagnostic challenge in infants receiving ARV prophylaxis or presumptive ART before definitive diagnosis.
Lack of evidence of ongoing HIV replication after eight years on ART
Children started on continuous ART in the CHER trial showed no sign of viral evolution after 7 to 8 years of treatment. 
In this study the investigators performed single-genome sequencing of the p6-PR-RT region of the HIV genome on plasma RNA before ART or PBMC DNA shortly after starting ART and on PBMC DNA approximately eight years after starting treatment. They compared HIV populations phylogenetically to look for emerging new variants. They also tested for panmixia to see if populations shifted over time. And they measured diversity of populations to see if there is accumulation over time.
Ten children were included: eight who started ART at less than 12 months and were fully suppressed on ART for 7 to 8 years and two replication controls who had viraemia for 1 to 3 years before or during ART.
The two controls showed clear evidence of HIV evolution. Data on one control with 1.3 years with detectable viral load out of 6.9 years on ART showed increased viral diversity (baseline 0.1%, long term 0.6%), a significant virus population shift by panmixia (p<10-4), and longer branches in ML trees.
The eight children with fully suppressed virus did diverge from founder virus. Data on one case who started ART at 1.8 years and remained on treatment for 8.1 years showed populations that were virtually identical, no panmixia shift (p=0.3) and no significant increase in viral diversity (baseline 0.04%, long term 0.1%). Results were similar among the eight children with suppressed virus.
“These data from early ART-treated children strongly refute the concepts that ongoing HIV replication is common on current ART regimens and that it replenishes the HIV reservoir”, the investigators wrote.
Nevirapine dosing regimen achieved target concentrations in HIV-exposed low birth weight infants
Data from IMPAACT P1106 showed nevirapine (NVP) dosed at 2 mg/kg once daily (birth to 14 days old) followed by 4 mg/kg daily achieved trough concentrations above the 0.1 ug/mL prophylaxis target in low birth weight infants <2500 g. 
IMPAACT P1106 is a Phase 4 study on PK and safety in low birth weight infants receiving antiretroviral and tuberculosis medicines as part of their clinical care in two South African sites.
Arm 1 looked at NVP HIV prophylaxis. Infants were stratified by birth weight: <1400 g (n=12), 1400 to <1800 g (n=12), and 1800 to <2500 g (n=16). PK samples and safety data were collected at study entry (day 7 to 14) and at 4, 6, 10, 16 and 24 weeks of age.
The study enrolled 40 low birth weight infants with mean birth weight of 1675 g (range 950 to 2460 g) and mean gestational age of 33 weeks (range 28 to 40).
There were 94 NVP trough concentrations available from 27 infants with mean weight of 2147 g (range 965 to 6050 g) and mean postmenstrual age of 37 weeks (range 29 to 56 weeks) at time of sampling.
The mean NVP trough concentration was: 1.87 ug/mL (range < 0.02 to 10.69); 6/94 (6%) < 0.1 μg/mL. Below target samples were all from later visits (median postmenstrual age 44 weeks; median weight of 3903 grams) when at home receiving NVP from caregiver.
At first visit, lower gestational age was associated with higher NVP concentration normalised for dose: r= -0.47, p=0.02. Across all visits, NVP trough concentrations were inversely related to infant postnatal age: r= -0.45, p<0.001).
Three infants died: two from sudden unexpected death and one from confirmed sepsis. Nine infants had Grade 3/4 expected AEs (common in premature infants), most frequent presumed or confirmed sepsis (n=6). Ten infants had Grade 3/4 unexpected AEs, most common being pneumonia (n=4). All AEs were unrelated to nevirapine.
Lopinavir/ritonavir super-boosting overcomes rifampicin interactions in children
Super-boosting LPV/r for a 1:1 ratio was safe and effective and overcomes rifampicin interaction for TB/HIV co-treated children in a South African study. 
This was an open-label, prospective, non-inferiority study evaluating super-boosted LPV/r (1:1) during rifampicin co-treatment compared with LPV/r (4:1) after stopping TB treatment in children weighing 3 to 15 kg.
Eighty of 96 enrolled children completed the study; 30% were <12 months at enrolment and seven completed the study before 12 months of age. TB treatment was started before ART in 73% children.
The percentage of modelled Cmin levels below target (<1 mg/L) was 7.6% (95% CI 0.4% to 16.2%) for super-boosting during rifampicin co-treatment, vs 8.8% (95%CI: 0.6% to 19.8%) without rifampicin. The median difference was -1.1% (95%CI: -6.9% to 3.2%), confirming the non-inferiority (10% threshold) of LPV exposure during super-boosting with rifampicin to standard LPV/r without rifampicin.
Lopinavir/ritonavir started at seven days of life impairs infant growth
The ANRS 12174 trial comparing LPV/r vs 3TC in HIV exposed uninfected children showed LPV/r started at seven days was associated with lower weight gain. 
In the trial, conducted in Burkina Faso, South Africa, Uganda and Zambia, 1273 HIV uninfected, breastfed children born to HIV positive mothers were randomised at seven days to either 3TC or LPV/r until the end of breastfeeding (maximum 50 weeks) as pre-exposure prophylaxis.
Infants were weighed and measured monthly and their z-scores calculated and compared. The analysis included 1266 children, with 14537 visits.
There was no difference in the height for age z-score between arms. But the weight for age score was lower in the LPV/r arm than in the 3TC arm: difference of means -0.22 (95% CI -0.34 to -0.09) p<0.01, at 26 weeks, and of -0.25 (95% CI -0.46 to -0.03), p=0.02, at 50 weeks.
The impact of LPV/r was greater for girls and, and in Burkina Faso and Uganda than in Zambia and South Africa.
The investigators will continue to follow up the children and look at whether or not this effect persists at five years old.
Targeted HIV screening at birth can identify most in utero transmissions
In utero transmission only occurred among infants identified as high risk in Botswana – using information available from the mother or her obstetric record at the time of delivery.  Targeting high risk infants will identify the large majority of in utero HIV transmissions
Botswana tests for in utero and intrapartum vertical transmission by infant HIV PCR at six weeks. The Early Infant Treatment Study was conducted to identify HIV infected infants at birth and offer immediate ART. Abstract eBook
Mothers were assessed for risk factors, which included: <8 weeks ART in pregnancy, last CD4 known <250 cells/mm3, last viral load >400 copies/mL, poor ART adherence in pregnancy, lack of maternal AZT in labour, and lack of infant post-exposure prophylaxis. Infants received heel stick and dried blood spots were collected for testing by PCR.
In the first year of the study, 4086 HIV exposed infants were delivered, 3541 (87%) had not been discharged, 2580 (63%) were eligible, and 2303 (56%) agreed to be screened for HIV.
Of those screened, 369 (16%) were identified as high risk for HIV. Twelve (0.5%) of the 2303 infants were identified as HIV positive at birth.
All 12 positive infants were identified as high risk at the time of screening, and all were identifiable as high risk by either: <8 weeks of maternal ART in pregnancy (9/157) or lack of maternal HIV suppression at last viral load test (3/6).
Unless stated otherwise, references are to the Programme and Abstracts of the 24th Conference on Retroviruses and Opportunistic Infections (CROI 2017), 13-16 February 2017, Seattle, Washington.
- Kuhn L et al. Treatment of acute HIV infection in neonates. CROI 2017, Seattle. Oral abstract 27.
- Veldsman KA et al. Rapid decline of total HIV DNA in children starting art within 8 days of birth. CROI 2017, Seattle. Oral abstract 28.
- MGK Katusiime et al. No evidence of ongoing HIV replication after 7 years on ART. CROI 2017, Seattle. Oral abstract 120.
- Bekker A et al. Pharmacokinetics of nevirapine prophylaxis in HIV-exposed low birth weight infants. CROI 2017, Seattle. Poster abstract 758.
- Rabie H et al. Lopinavir/ritonavir 1:1 super-boosting overcomes rifampicin interactions in children. CROI 2017, Seattle. Oral abstract 29LB.
- Nagot N et al. Lopinavir/ritonavir initiated at 7 days of life impairs infant growth. CROI 2017, Seattle. Poster abstract 781.
- Ibrahim MR et al. Targeted HIV screening at birth can identify the majority of in utero transmissions. CROI 2017, Seattle. Poster abstract 790.